Josef Šivic receives ERC Advanced Grant to investigate intelligent systems capable of learning in a dynamic 3D world

0
2427

Dr. Josef Šivic from the Czech Institute of Informatics, Robotics and Cybernetics (CIIRC) of the Czech Technical University in Prague was awarded an ERC Advanced grant to conduct research on intelligent systems capable of learning in dynamic 3D environments. The project FRONTIER: Federated foundational models for embodied perception, will develop a new generation of large-scale neural models for machines that learn in a dynamic 3D world and interact with their surrounding environment. The recipients of this highly competitive grant, totalling up to €2.5 million over five years, have been announced today by the European Research Council (ERC). Josef Šivic is ranked by Research.com as one of the top computer scientists in the Czech Republic to which he returned from France thanks to his previous European-funded research project. 

The FRONTIER project combines computer vision, machine learning and robotics and aims to develop new neural architectures that combine large-scale neural networks with learnable physical simulation components. The latest artificial intelligence systems show remarkable performance on tasks considered challenging by many people, such as writing computer programs or creating images. However, other tasks that humans find easy and perform every day, such as perceiving and interacting with the surrounding environment, are still difficult for machines.  For example, commonly used industrial robots are pre-programmed for one specific task but are unable to easily perform other tasks in the changing three-dimensional world. And this is also the challenge addressed by Josef Šivic and his team.

In their research, they will design new algorithms that will enable machines to learn complex multi-step tasks with the aim of bringing them closer to humans who can often learn from only a few available examples. The most ambitious goal is to develop new methods to allow sharing and accumulating learning experiences across different systems thereby achieving new levels of scale, accuracy and robustness not achievable by learning in any individual system alone. This is a challenging problem as individual systems are performing different tasks in very different environments.

The next challenge is to come up with large-scale neural network models for embodied perception, which take into account the physical and geometrical structure of the world to achieve generalization across tasks, situations, and environments”,  says Dr. Josef Šivic and also explains the potential applications: „Examples of such embodied systems include a robot as a home assistant, a robot in a factory, a self-driving car, or even a robotic surgeon. All of these systems have some kind of a physical body, need to perceive the world around them, reason about it, and interact with it. However, current large neural models have only limited capabilities in this area.

Breakthrough progress on these problems would have a major impact on our daily lives. For example, the findings from this research could lead to safer cars that learn from each other, intelligent production lines that collaboratively adapt to new workflows or a new generation of intelligent assistive robots that automatically learn new skills from the Internet and each other.

„The competition for ERC grants is incredibly fierce, with applications being reviewed by panels of top scientists,“ said Professor Vladimír Mařík, Scientific Director of CIIRC. „At our institute, we strive to create an environment that enables our researchers to conduct truly groundbreaking research, and I am delighted that Dr. Šivic will be conducting his FRONTIER project here and with our full support. It is also important for Czech science to have extraordinary scientists working here. 

The European Research Council (ERC) has announced the awarding of 218 Advanced Grants to outstanding research leaders across Europe, as part of the Horizon Europe programme on Thursday, March 30 2023. The grants – totalling €544 million – support cutting-edge research in a wide range of fields, from medicine and physics to social sciences and humanities. This year 4 Advanced Grants are hosted in the Czech Republic. The evaluation panels assessed 1,650 applications with an overall success rate of 13.2%, with the Czech Republic being less successful than other countries in obtaining these grants in the long term.

 

Further information

Dr. Josef Šivic is the head of the Intelligent Machine Perception research group at the Czech Institute of Informatics, Robotics and Cybernetics at the Czech Technical University in Prague and the director of the ELLIS Unit Prague. Since 2022, he is also the chairman of the AICzechia initiative, which brings together, via their leading researchers in artificial intelligence, forty academic teams and departments, mainly at universities and research institutions in the Czech Republic. His research group also contributes to the ranking of CTU among the top five European institutions in the field of computer vision (CSRankings for the period 2018 – 2023), and he is ranked among the best Czech computer scientists by Research.com (More information here). 

After completing his masters degree at CTU in 2002, he spent more than 15 years abroad, including a PhD at the University of Oxford, UK, a postdoctoral fellowship at the Massachusetts Institute of Technology, USA, and more than a decade at the French National Institute for Research in Digital Science and Technology (Inria) in Paris, France, where he was awarded an ERC Starting Grant. Since 2017, he has been working at the CIIRC CTU, where has led the 111 million CZK OP R&D IMPACT project that enabled his return to the Czech Republic. He has co-authored more than 100 scientific publications and was awarded the Helmholtz Prize and the Longuet-Higgins Prize for fundamental contributions to computer vision.

 

ERC – European Research Council set up by the European Union in 2007, is the premier European funding organisation for excellent frontier research. It funds creative researchers of any nationality and age, to run projects based across Europe. The ERC offers four core grant schemes: Starting Grants, Consolidator Grants, Advanced Grants and Synergy Grants. With its additional Proof of Concept Grant scheme, the ERC helps grantees to bridge the gap between their pioneering research and early phases of its commercialisation. The ERC is led by an independent governing body, the Scientific Council. Since November 2021, Maria Leptin is the President of the ERC. The overall ERC budget from 2021 to 2027 is more than €16 billion, as part of the Horizon Europe programme, under the responsibility of the European Commissioner for Innovation, Research, Culture, Education and Youth, Mariya Gabriel.

Previous articleThe DFLEX project verified the usability of flexibility aggregation
Next articleCzech success in the prestigious grant scheme to support top basic research: Josef Šivic, holder of ERC Advanced Grant 2022